Message Transmission Algorithm Based On Community Connectivity In Opportunistic Networks

Now with the prevalence of the short distance devices, such as Wife, Bluetooth,ZigBee and other wireless communication，it proper the development of opportunisticnetworks. Since the nodes move anywhere, the wireless communication distance andhardware resource of nodes are limited, the network topology changes anytime, thenetwork shows local connectivity and global intermittent connectivity. The traditionalpacket forwarding mode can not be implemented in this kind of network, so the dataforwarding of opportunistic networks face enormous challenges. How to improve theperformance of network and the success rate of data forwarding and reduce thetransmission delay of message and reduce the load of network is problem. In this paper,according to activities of humans have the nature of community. Message transmissionalgorithm based on community connectivity in opportunistic networks was proposed.This paper’s main works are followed:(1)This paper discussed the community partition. Based on the characteristics ofpeople who have similar features more like live together. So this paper set some fixednodes to mark the communities and to determine the node to the community to providelogo. After a period of time, each node chooses itself community where it stays longesttime.(2) Community connectivity was proposed. According to the table of nodes historygoing, it can calculate the connected relation between communities and get thecommunity connectivity-the greater is more frequently between communities.(3)Message transmission algorithm based on community connectivity inopportunistic networks was proposed. It is divided into two kinds of transmissionalgorithm, message within the community and message between communities. Themessage between communities is based on the community connectivity. it can find atransmission path with the highest probability to transport the message. The messagewithin the community is based on the nodes of historical encounter probability and theactivity of nodes.(4) The simulation from two aspects in the paper, including the time-to-live ofmessage and the buffer size of nodes. Compared with the Spay and Wait algorithm, theexperiments show that this algorithm increases the message transmission rate andreduces the average message transmission delay and the message storage time.